[attachment=13313]
Abstract
The Seminar study focusses on Power Controlled MAC layer in Ad Hoc Networks. A
detailed study of the various challenges involved for deploying power control ad hoc networks
is carried out. Methods to overcome the problems are studied. This seminar provides a
comparative study of the different approaches for power control to achieve a high channel
utilization and energy saving.
1 Introduction
A Mobile Ad hoc network (MANET) is a multihop wireless network where a group of nodes co-operate to maintain connectivity. Each node in the network also performs routing functions
in a MANET. Such networks are infrastructure less and are deployed in places where it is not
feasible to install the desired infrastructure in a short period of time. The main issues in these
highly mobile networks are maximum channel utilization, power conservation, maintaining connectivity.
As advanced applications increase, there is requirement for more bandwidth. Power
Control in MANET helps to increase spatial reuse and hence improves the overall channel utilization. Multiple access protocols and power control provide better channel utilization. The
energy saved provides long lifetime of the nodes resulting in a better network connectivity. Better
channel utilization provides overall increased throughput.
Efficient channel allocation and management is crucial for better performance of the network.
Principle method to manage resources is to share them. Sharing is handled by classical
approaches like Frequency Division Multiple Access (FDMA), Time Division Multiple Access
(TDMA), Code Division Multiple Access (CDMA). TDMA uses time slots separated in time to
access resources. FDMA uses on-overlapping frequency bands for access. CDMA uses different
encoding codes to separate multiple access. These schemes use a either use a fixed transmission
power level or there exists a central authority to decide the appropriate power level for the transmission.
A fully distributed scheme is difficult to manage. The work discussed in this seminars
is related to distributed schemes to share channel among various mobile nodes in MANET.
The 802.11, the de facto standard that is used in wireless networks, is based on CSMA/CA
paradigm. In 802.11 the RTS-CTS exchange is used to reserve transmission floor for the packets
to be transmitted. This collision avoidance scheme causes the potentially interfering nodes to
defer transmission and wait for the channel to be free. The work in this area has been focussed
to improve this protocol for optimum power utilization and channel reuse. Researchers have
been working to preserve the collision avoidance property of 802.11 while trying to achieve interference
limited transmissions to occur simultaneously. The simultaneous transmissions use
variable power such that it does not affect any ongoing transmission. This allows for a overlap of
the transmission floor and hence better channel utilization and overall average throughput. This
is adopted in approaches covered in [2] [3] [4]. Approaches like [5] try to achieve a throughput
equivalent to 802.11 but with a lower energy consumption.
The various schemes discussed are used for power control for efficient energy consumption and
better channel utilization. Power Control for Medium Access Control [2] generalizes the “onoff”
collision avoidance mechanism to a “variable bounded power” transmission. The sourcedestination
pairs are more tightly coupled allowing more simultaneous transmissions. In CDMA based scheme is used to facilitate multiple transmissions. In addition to the energy
savings, power control is used to overcome many problems related to non-orthogonal codes. The
approach in [4] uses channel gain information from overheard RTS-CTS packets to dynamically
construct the network topology and perform interference-limited simultaneous transmissions.
The power controlled MAC protocol [5] tries to optimize the power consumption of the 802.11
while achieving the same throughput. In this scheme, data is transmitted at the negotiated
lower power, but at periodic intervals the data is transmitted at a higher power to facilitate
detection by the other nodes.
The report is organized as follows. Chapter 2 explains the problems associated with the 802.11
based networks which form the basis of the work on Power controlled MAC layer for Ad Hoc
Networks. Chapter 3 introduces a few challenges involved in solving the power control problem
in Ad Hoc Networks and explains the non-triviality of the solution. Chapter 4 gives a common
framework which is followed to achieve a power controlled MAC and the assumptions made are
explained. Chapter 5 discusses the various solution approaches in detail. Chapter 6 discusses the
special issues that need to be handled and the methods proposed for them. Chapter 7 provides
concluding remarks for the seminar study.
2 Problem with 802.11 Networks
The basic 802.11 MAC protocol that is used as a de facto standard in Ad Hoc Networks is
inefficient in terms of energy consumption and channel reuse. This section explains problems
associated with using the basic MAC protocol. The main problem areas include the fixed single
power transmission and the on-off operation of RTS-CTS.
All data is transmitted with a constant power in 802.11, this leads to a wastage of power.
When two nearby nodes wish to communicate more power is used, while much less power can
be sufficient. The basic 802.11 scheme supports only constant power transmission. This results
in a more transmission floor reserved for the communication. This means that nodes in a larger
area are made to defer their communication till the current transfer is complete. This leads to
wastage of time and bandwidth, which is a scarce resource in wireless medium. RTS-CTS is
used by 802.11 to eliminate Hidden terminal and exposed terminal problems. When RTS-CTS
messages are exchanged between two nodes for data transfer the neighboring nodes that intend
to transmit data defer their transmission till the time the current transmission is complete. This
simple turn-off on part of the neighbor leads to wastage